Optimization of mask size with median modified Wiener filter algorithm for gamma images using pixelated semiconductor detector: Monte Carlo simulation study

Abstract Gamma images inherently contain noise distributions owing to the low photon count in a detector. Several methods using semiconductor detectors have been reported to overcome this challenge. Median modified Wiener filter (MMWF) algorithm is one of the effective noise reduction methods, which employs mask size as an essential parameter for optimization. In this study, we evaluated and optimized the acquired gamma image using the proposed MMWF algorithm based on the Monte Carlo simulation tool with various mask sizes (3 × 3 , 5 × 5, and 7 × 7) in a cadmium telluride pixelated semiconductor detector. Our results demonstrated that the images obtained with 7 × 7 mask size showed the lowest values in the normalized noise power spectrum. The contrast-to-noise ratio and the coefficient of variation were observed to be 29.2–65.9% better for the images obtained with the 7 × 7 mask size compared with those obtained using other mask sizes. This study suggests that an excellent image quality of gamma images can be achieved with semiconductor detectors by optimizing the mask size in the MMWF algorithm.